Abstract
Within the past decade, considerable numbers of drugs for the treatment of diabetes and metabolic disorders have either not reached the market, having been stopped late in development, or having been withdrawn after initial approval soon after market authorization due to serious safety concerns. Many formerly promising drug candidates with block buster potential in their respective therapeutic area have foundered, often with high financial losses for the developing drug company (Table 1). A critical look back over the past 10 years leads me sceptically to ask if we are currently experiencing a closing sale of innovative treatment options for treating diabetes and metabolic disorders?
Drug non-approvals and early withdrawals in the past decade for the treatment of T2DM and metabolic disorders
Type 2 diabetes mellitus. ** High-density lipoprotein.
This inglorious round dance of development failures and market withdrawals in the past decade started with the thiazolidinedione troglitazone, which was stopped in clinical development due to increased liver toxicity [Smith, 2003; Pishvaian et al. 2004]. In August 2001, the cholesterol- lowering drug cerivastatin was withdrawn from the market due to increased myotoxicity and an increased number of reports about rhabdomyolysis including outcomes with death [Thompson et al. 2003]. It is at least debatable if cerivastatin was so much more dangerous than currently still available statins and if it was justified to pull it from the market, rather than reducing the available dosage and increasing the awareness in prescribers of critical drug interactions such as that with gemfibrozil.
As insulin resistance and dyslipidemia are generally recognised as early offenders in the process of developing type 2 diabetes mellitus (T2DM), the concept of the dual peroxisome proliferator-activated receptor (PPAR) α/γ -agonists – so-called ‘glitazars’ - was born [Bailey, 2007]. These dual-action molecules combined the antihyperglycaemic and insulin-sensitizing properties of PPARγ -agonism with triglyceride-lowering and high-density lipoprotein cholesterol HDL-raising effects. Tesa- and Muraglitazar proceeded through phase III trials and showed efficacy against the hyperglycaemia and hypertriglyceridaemic dyslipidemia of T2DM, however neither entered the market, as both had unacceptable side effects. For muraglitazar, the US FDA required further safety studies due to the incidence of oedema, heart failure and cardiovascular deaths during the trials. For tesaglitazar there were concerns about elevated serum creatinine and decreased glomerular filtration. From the glitazar-family, only aleglitazar survived and is still in the race and the favourable results of the short-term study SYNCHRONY over 16 weeks have been recently published [Henry et al. 2009]. It is currently in phase III studies for patients with T2DM and coronary heart disease for the reduction of cardiovascular mortality and one of the last really promising molecules in the future if the adverse-event profile promises a suitable benefit:risk ratio.
The cholesterol-ester-transfer-protein (CETP)-inhibitor torcetrapib was a drug being developed to treat hypercholesterolemia and to prevent cardiovascular disease. The drug acts by inhibiting cholesteryl ester transfer protein (CETP), which normally transfers cholesterol from HDL cholesterol to very low density or low density lipoproteins (VLDL or LDL). Inhibition of this process results in higher HDL levels, the ‘good’ cholesterol- containing particle and reduces LDL levels, the ‘bad’ cholesterol. Its development was halted in 2006 when phase III studies showed excessive all-cause mortality in the treatment group receiving a combination of atorvastatin and the study drug.
Between September 2006 and October 2007, a new method of delivering insulin to the body was available for patients. This inhalable drug form of insulin was marketed as Exubera and was considered short- or rapid-acting insulin. In clinical studies, Exubera reached peak concentration levels faster than some insulins administered by injection. Thus, this form of insulin would begin working within the body faster than those forms of injected insulin. However, Type 1 and 2 diabetics still needed an additional injection of longer acting insulin to maintain a basal level for a 24 hour period. In April 2008, Pfizer announced that of the 4,740 patients who used Exubera in clinical trials, six had developed lung cancer as of April 2008, compared with only one of the 4,292 patients in the placebo group. The drug was withdrawn from the market in October 2007. The official reason for stopping the marketing was that the drug ‘failed to gain acceptance among patients and physicians.’ Unofficially it became obvious that the drug was stopped mainly because of unexpectedly low sales. After the withdrawal of the only inhalable formulation, all currently available insulin formulations are administered by subcutaneous or intravenous injection.
The next new developed drug which came under fire was the selective CB1-receptor-antagonist rimonabant, which was licensed by the EMA for the treatment of obesity. Rimonabant has been authorized in the EU since June 2006 as an adjunct to diet and exercise for the treatment of obese patients or overweight patients with associated risk factors, but has never been launched in the US or Canada. Warnings about psychiatric side effects, in particular depression, have been included in the product information since the drug was first authorized and these have been continuously updated and strengthened to include further contraindications and upgraded warnings about the risks associated with the use of this drug. The CHMP concluded that there is an approximate doubling of the risk of psychiatric disorders in obese or overweight patients taking rimonabant compared with those taking placebo. The market authorization in the EU was thereafter suspended in 2008. Rimonabant was never approved by the FDA.
The weight loss drug sibutramine was not approved by the FDA in February 2010 due to increased cardiovascular safety concerns. The Sibutramine Cardiovascular Outcome Study (SCOUT) investigated the influence of sibutramine-induced weight loss on cardiovascular endpoints because it was assumed that especially patients with high cardiovascular risk might benefit from sibutramine-induced weight loss. In fact, patients with preexisting cardiovascular conditions who were receiving long-term sibutramine treatment had an increased risk of nonfatal myocardial infarction and nonfatal stroke [James et al. 2010].
A smouldering drug safety concern that attracted our attention for the first time arose from a metaanalysis published by Steven Nissen in 2007, who revealed an increased cardiovascular risk for patients taking rosiglitazone [Nissen et al. 2007]. It finally came to an end with the suspension of the marketing authorizations for all rosiglitazone-containing anti-diabetes medications licensed in the EU in late 2010 by the EMA. However, the US FDA has decided that rosiglitazone can remain available, but only under a very stringent restricted-access program. Remarkably, post-hoc analyses did not reveal an increased risk for macrovascular events with pioglitazone [Erdmann et al. 2010]. For a little while it looked as if pioglitazone might be the glitazone of choice offering the therapeutic benefits of lowering blood glucose and insulin resistance without increasing CV risk. However, this hope lasted only briefly and in July 2011, the worldwide diabetes community was again alarmed by the EMA, which announced a small increased risk of bladder cancer with pioglitazone, triggered by an interim report of an ongoing cohort study examining the association between pioglitazone therapy and the risk of bladder cancer in patients with diabetes [Lewis et al. 2011]. This decision came one month after an announcement by the FDA that it, too, was revising labeling and patient information. French regulators opted to suspend sales of the drug, while German regulators announced that pioglitazone should not be started in new patients.
Unfortunately, this is still not the end of diabetes drug safety concerns. It is well known that the risk of developing a range of solid tumours such as pancreatic cancer [Magruder et al. 2011] is increased in type 2 diabetes, and may be influenced by glucose-lowering therapies. A number of observational studies have linked insulin glargine (A21Gly,B31Arg,B32Arg human insulin) with a putative increased cancer risk, and particularly with breast cancer [Renehan, 2011; Ruiter et al. 2011]. The results of these studies caused considerable uncertainty about drug safety, not only among patients on insulin analogues, but also among physicians prescribing these drugs. However, a retrospective cohort study examining the risk of development of solid tumors in relation to treatment with oral agents, human insulin and insulin analogues revealed that patients on insulin or insulin secretagogues were more likely to develop solid cancers than those on metformin, but combination with metformin abolished most of this excess risk. Metformin use was associated with lower risk of cancer of the colon or pancreas, but did not affect the risk of breast or prostate cancer. Use of insulin analogues was not associated with increased cancer risk as compared with human insulin [Currie et al. 2009]. In 31 randomized controlled trials, insulin glargine was not associated with an increased incidence of cancer, including breast cancer, compared with the comparator group [Home et al. 2009]. Taking together this currently available evidence, it seems unlikely that current data really allow the conclusion that treatment with insulin glargine is associated with increased cancer risk [Mussig et al. 2011].
Early in 2011, even the newest classes of antihyperglycemic drugs, GLP-agonists and DPP-4-inhibitors, came under pressure: An analysis of the US FDA’s database of reported adverse events for those associated with the dipeptidyl peptidase-4 inhibitor sitagliptin and the glucagon-like peptide-1 mimetic exenatide investigated rates of reported pancreatitis, pancreatic and thyroid cancer, and all cancers associated with sitagliptin or exenatide, compared with other therapies. This analysis concluded that pancreatic cancer was more commonly reported among patients who took sitagliptin or exenatide as compared with other therapies and found an increased risk for pancreatitis with glucagon-like peptide-1-based therapy [Elashoff et al. 2011].
The most recent embarrassing news in the diabetes drug session is the negative decision on the prototype molecule of the new group of SGLT-2-inhibitors dapagliflocin. In July, the FDA advisory committee voted 9:6 against the approval of this new drug due to increased incidence of liver toxicity and cancer. Dapagliflozin is the most advanced candidate in a new class of drugs known as sodium-glucose co-transporter-2 (SGLT-2) inhibitors. Such drugs are designed to lower blood glucose levels in patients with diabetes by increasing the amount of glucose excreted in the urine. The SGLT-2 system is used by the kidneys to filter and reabsorb glucose. The FDA found dapagliflozin to be effective at lowering blood sugar levels but was concerned about several unexpected safety issues including an imbalance in breast and bladder cancers as well as one possible case of liver injury that might be drug-related. Late in October 2011 the FDA confirmed that it was extending its review on dapagliflocin, delaying a decision on whether to approve the drug for adults with type 2 diabetes to January 28th 2012. The agency requested additional information from recently completed and ongoing late-stage trials of the therapy, which the companies developing dapagliflocin are submitting, and the regulator will use the next three months to review the data.
What have we learned from the drug development failures of the past?
Doesn’t this obviously never ending story of development failures and upcoming safety concerns with diabetes and anti-obesity drugs look like a closing sale of innovative treatment options for type 2 diabetes and obesity? Both diseases are, besides cancer, undoubtedly the largest upcoming health threats especially in developed, but also in developing-world countries in the 21st century. Can we really afford to frivolously give up promising and effective treatment options in the fight against T2DM due to sometimes obviously more than debatable drug safety concerns? Moreover, many individuals with prediabetes already show diabetes-related complications [Fox et al. 2005]. In the DECODE-study, impaired glucose tolerance was associated with a significantly increased all-cause-mortality [Group, 1999] illustrating that already people with prediabetes represent a high risk population for cardiovascular and diabetes-related complications. Diabetes strongly influences development and progression of cardiovascular disease [Schindler, 2010]. This rather supports earlier drug treatment in the course of the disease [Hanefeld et al. 2011].
UKPDS taught us that blood sugar reduction reduced microvascular complications by 25% [Group, 1998]. This result let us to assume that blood sugar reduction must necessarily also reduce macrovascular complications. Indeed, at the end of the intervention in 1997, UKPDS did not show a significant reduction of macrovascular complications. Also, the ACCORD- and ADVANCE-studies did not reveal a macrovascular benefit induced by HbA1c-reduction. The rosiglitazone story definitely taught us that the apparently impeachable evidence-based mantra of diabetology since the UKPDS that drugs reducing blood sugar must also reduce macrovascular complications needs to be scrutinized. Since the rosiglitazone experience it is obvious that measuring only surrogate endpoints such as HbA1c is not sufficient to evaluate clinical value and safety profile of a new drug. The FDA reacted in 2008 with a new guideline for industry which now requires conduction of a metaanalysis after termination of the clinical study program of a new drug analysing the MACE- (major adverse cardiac event) liability before a new drug application can be filed. An upper CI ≥ 1.8 indicates hereby an unacceptable CV risk and would require a large additional safety study. A CI ≥ 1.3 and < 1.8 indicates positive benefit:risk ratio but would require an additional post-marketing CV safety trial for confirmation. In the case where the CI lies < 1.3, the drug can be approved without additional post marketing safety trial.
Perspectives
What we need in the future is a sustainable concept for effective risk management in order to guarantee a maximum available drug safety. How can we anticipate drug safety problems and, even more important, how can we differentiate between adverse events definitely caused by a drug from incidences of naturally occurring diseases? Looking back, there is no doubt that it was not useful to severely unsettle patients and prescribing physicians with preliminary and non-validated safety concerns and potential risks, for example with the insulin-analogue insulin glargin, GLP-1-agonists and DPP-4 inhibitors.
We do not need larger or longer preregistration studies, but we do need a better system of ongoing vigilance once an agent reaches the public, possibly mediated by provisional registration, and supported by mandatory independent pharmacovigilance [Gale, 2009] also involving renowned drug and diabetes experts from academia. We also need greater transparency: The webpages of EMA and FDA are still very difficult to navigate and it is still difficult, even for experts, to access the relevant material. This should be changed soon.
There is no doubt that effective new treatment strategies should be pharmacologically as much selective as possible. DPP-4-inhibition for example is able to effectively lower blood glucose levels and HbA1c without increasing the risk for severe hypoglycemia. This therapeutic value is unique and well suited especially for combination treatment. However, we should not forget that the enzyme DPP-4 is ubiquitously available in the human body. This means that we may not completely exclude adverse events triggered by enzyme-inhibition in anatomic non-target structures. This consideration and the experience with rosiglitazone underline the necessity of long-term safety studies, which are already initiated (SAVOR-TIMI (Saxagliptin), TECOS (Sitagliptin), EXAMINE (Alogliptin) and CAROLINA (Linagliptin)). The gliptins already provide a large amount of safety data today from pooled analyses including 24000 patients in completed clinical trials covering almost 12500 subject-years of exposure solely with sitagliptin and vildagliptin [Williams-Herman et al. 2010; Ligueros-Saylan M et al. 2010]. Nonetheless, critically evaluating the current available evidence today does not justify any changes in the prescribing behaviour of GLP-1-agonists and DPP-4-inhibitors.
Aleglitazar, a combined PPARα and γ-agonist, which is the only combined PPAR-agonist that is still in development, shifts the body’s energy substrate preference from glucose to lipids and hence contributes to the reduction of blood glucose level [Adeghate et al. 2011]. Dual or pan PPAR-ligands stimulate two or more isoforms of PPAR and thereby reduce insulin resistance and prevent short- and long-term complications of diabetes including micro-and macroangiopathy and atherosclerosis, which are caused by deposition of cholesterol. The use and testing of many dual/ pan PPAR modulators have been discontinued because of severe side effects including renal failure, fibrosarcomas, urinary cancer and anemia, the main achilles heel of this treatment. The search for an endocrine ‘master switch’ determining the metabolic susceptibility and underlying the clinical manifestation of the metabolic syndrome (MetS) is still ongoing. There is considerable evidence supporting an interpretation of the MetS as an endocrine disease and RAS activation, obesity and insulin resistance seem to be mainly related to the pathogenesis of the MetS, whereas lipid abnormalities, hypertension and atherosclerosis might be rather comorbidities [Schindler, 2007]. Dual PPAR-stimulation is definitely still a very interesting pharmacological approach for causally targeting the MetS in the near future because this treatment principle has been suggested to uniquely target multiple components of the metabolic syndrome, including obesity, insulin resistance, hyperglycemia, dyslipidemia and atherosclerosis. The future will show us if the benefits overweigh the risks.
The newest antidiabetic drug treatment principle, SGLT-2 inhibition, is still under investigation by the regulatory authorities. Hopefully the detailed data review on dapagliflocin as prototype molecule will finally not show an increased safety risk of the class. This treatment principle is pharmacologically selective. By dumping calories it targets obesity, which is one of the most important components of the pathogenesis of T2DM. Therefore, SGLT-2 inhibitors may have indications both in the prevention and treatment of T2DM, and perhaps T1DM, with a possible application in obesity. They might also be well suited as add-on-therapy if monotherapy is not sufficient to control blood glucose. Additional diuretic and hypotensive effects have been suggested and are still under investigation in clinical trials.
Conclusions
Pharmacological intervention of T2DM should target the pathophysiology of the disease via different molecular targets and as selectively as possible, which allows combination treatment with different glucose-lowering drugs in low doses. Drug development should be guided by this principle in the future. Better and more individual-patient screening prior to initiation of antidiabetic drug therapy might additionally help to minimize adverse events by selecting the most specific appropriate pharmacological intervention for the individual patient. We should also not forget that besides HbA1c-reduction, pleiotropic effects of antidiabetic drugs might have positive impact on the MetS and on cardiovascular endpoints.
Looking back to the development failures and early withdrawals of the past 10 years and to the latest drug safety discussions in diabetic and metabolic medicine, I hereby enduringly proclaim that we must stop the uncritical and unfounded closing-sale of innovative drug treatment options now. Otherwise there will be only blunt weapons left for the fight against the upcoming epidemic of obesity and T2DM in the 21st century.